BALANCE CONFIDENCE AND FUNCTIONAL BALANCE ARE 1 ASSOCIATED WITH PHYSICAL DISABILITY AFTER HIP FRACTURE

ABSTRACT


INTRODUCTION 1
Hip fracture is a common and severe trauma in older people, leading to balance 2 impairments, decreased muscle strength, and loss of independence in daily activities [1,2]. 3 It's also known that only half of the hip fracture patients regains their pre-fracture level of 4 functional ability [2] and 15% will be institutionalized permanently [3]. Long-lasting 5 mobility limitation after hip fracture may lead to prolonged physical disability as well as 6 new injurious falls and fractures [2]. The incidence of hip fractures increases with age and 7 the total number of fractures is expected to rise due to ageing of the population [4]. 8 9 Although only five percent of all falls cause a fracture, approximately 95% of all hip 10 fractures are caused by a fall [5,6]. The risk factors for falls interact with each other and 11 large individual variation exists. The risk of falling increases rapidly with the number of 12 risk factors. Impaired functional balance is considered the most common risk factor for 13 further falls and fractures after a previous hip fracture [7]. Balance control has also a 14 fundamental role in various activities of daily living, especially in those that require 15 independent standing or walking. Furthermore, those who have had a fall with a traumatic 16 consequence often experience fear of falling even years after the incident [7]. Additionally, 17 fear of falling has been strongly associated with future falls [8,9]. 18 19 One method of operationalizing fear of falling is to assess self-reported balance confidence 20 using the Activities-specific Balance Confidence Scale (ABC) described by Powell et al. 21 [10]. Decreased balance confidence has been associated with poor functional balance, 22 increased disability, and reduced quality of life in community-dwelling older people 23 [8,11,12]. Individuals with low balance confidence and balance impairments are also likely 24 to reduce their physical and social activity, which in turn predicts the onset of disability 25 [13]. Based on our clinical experience, self efficacy and balance confidence are low in hip 1 fracture patients who have suffered a traumatic fall accident, which might drastically slow 2 down or disable the rehabilitation process. At the moment the scientific evidence regarding 3 the association between balance confidence and rehabilitation outcome after hip fracture is 4 insufficient. 5 6 To our knowledge there are no other studies that have simultaneously examined 7 associations of low balance confidence and balance impairments in relation to physical 8 disability, in older people who have suffered a hip fracture. However, these associations 9 should be studied to better understand the factors potentially affecting the recovery and 10 rehabilitation processes after hip fracture. The purpose of this study was to investigate the 11 associations between decreased balance confidence, impaired functional balance, and 12 physical disability among older people who have sustained a previous hip fracture. ambulatory and community-dwelling people who were living in the city of Jyväskylä or 24 the neighboring municipalities, and had been operated for femoral neck or trochanteric fracture (ICD code S72.0 or S72.1). All potential participants (n=748) were informed of 1 the study by a written information letter. Those willing to participate (n=293) were 2 interviewed over the telephone or met during the inpatient period at the health care centre 3 to ensure their suitability for the study. The exclusion criteria were: inability to move 4 outdoors without assistance of another person, amputation of a lower limb, severe 5 progressive or neurological diseases, alcoholism and severe memory problems

Review of the medical data and health status 17
During a medical examination performed by a nurse and physician, the presence of chronic 18 conditions, the use of prescription medication, fracture status, and the date and type of 19 surgery were confirmed according to a pre-structured questionnaire, current prescriptions, 20 and medical records obtained from the local hospital and health care centers. 21 Contraindications for participation in muscle strength and balance assessments were 22 evaluated by the physician [17]. 23

Balance confidence 1
A modified Finnish version of the Activities-specific Balance Confidence Scale (ABC 2 [10,18]) was used to assess confidence in performing specific activities without becoming 3 unsteady. Balance confidence can be regarded as a measure of fear of falling [10]. The 4 modified ABC scale consists of 16 items. Subjects are requested to describe, how 5 confident they are in carrying out different tasks, indoors and outdoors. Answers for each 6 question were rated from 1 (no confidence) to 10 points (total confidence). The total score 7 ranges from 16 to 160 and higher scores indicate better balance confidence. 8 9

Functional balance 10
Functional balance was assessed by the Berg Balance Scale (BBS [19]) which evaluates 11 the ability to perform 14 different tasks such as standing up, sitting down, reaching and 12 turning around oneself, looking over the shoulders and standing on one foot. The ability to 13 perform each task is rated from 0 (incapable) to 4 (safe and independent). The total score 14 ranges between 0-56 and higher scores indicate better functional balance. food, doing laundry, coping with light house work, coping with heavy house work, 22 handling medication, using the telephone, using public transportation, and handling 23 finances [20,22]). There were five response categories: 1) I manage without difficulties, 2) 24 with some difficulties, 3) with lots of difficulties, 4) I can't manage without assistance of another person, and 5) I can't manage even when assisted. The original categorical 1 variables were dichotomized; a) Without difficulty (category 1) and b) Difficulty 2 (categories 2-5). Subsequently, two sum scores were composed: ADL score (ranging from 3 0-6) and IADL score (ranging from 0-8). Higher scores indicate more difficulty. 4 5

Confounders 6
Isometric muscle force (Newton) for knee extension was measured on the fractured side by 7 an adjustable dynamometer chair (Metitur Ltd, [23]. During the measurement the ankle 8 was attached to a strain-gauge system with the knee angle fixed at 60° from full extension. 9 Participants were encouraged to extend the leg as hard as possible. After two to three 10 practice trials, measurements were performed at least three times until no further 11 improvement occurred. Each contraction was maintained for two to three seconds. The 12 inter-trial rest period was 30 seconds. The performance of the highest maximal force was 13 used for analysis. 14 15 Pain on the fractured side was assessed by two questions "Have you experienced pain in 16 the lower back, hip, knee, ankle or foot on your left/right side? Has the pain compromised 17 your mobility?" The response alternatives were 1) No, 2) Yes, but it is not offending, 3) 18 Yes, and it is offending. A new variable "offending pain of the fractured side" was 19 composed based on the answers. The use of walking aids outdoors was assessed by the 20 question: "Do you use walking aids when going outdoors? Response alternatives were 21 Yes/No. 22

Statistical analysis 1
The means, standard deviations, frequencies, and percentage values were calculated for the 2 background variables. The associations between balance confidence and physical disability 3 as well as between functional balance and physical disability were assessed by negative 4 binomial regression which is a generalization of the Poisson regression that accounts for 5 the over dispersion detected in Poisson models. The negative binomial regression model 6 takes into account that disability tends to be a cumulative phenomenon and that having 7 difficulty in one activity makes it more likely to have difficulty in two or more activities. The Kolmogorov-Smirnov test was used to test the normality of distributions. Only the 20 ABC score was normally distributed. Thus, the Spearman correlation was used to analyze 21 the association between the ABC and BBS. The Spearman ρ was raised to the second 22 power to express the coefficient of determination (R 2 ) for ABC relative to BBS. 23 Regression modeling was performed using STATA 12 statistical software. All other 24 analyses were performed using PASW Statistics 18.

RESULTS 1
The average age of the participants was 77.4 (SD=7.2) years and the mean time elapsed 2 since the fracture was 1.7 (2.1) years. Seventy-three percent was female, 47 % had an 3 internal fixation and 53 % an arthroplasty operation. The mean ABC score was 91.5 (32.3) 4 points, i.e. 58 % of the maximum score. Furthermore, the mean BBS score was 44.1 (9.3) 5 points, i.e. 79 % of the maximum. The median values of ADL and IADL score were one 6 and three, respectively. The participant characteristics are presented in Table 1 The ABC and BBS scores correlated highly but not fully (Spearman ρ=0.69) and the ABC 23 score explained 48% of the variation in BBS score (R 2 =0.476). However, placing the ABC 24 and BBS scores together in the same regression model with ADL or IADL did not materially change their individual IRR's ( Table 2). This implies that the ABC and BBS 1 represent partly different phenomena and they cannot be considered as surrogate 2 measurements. However, in the fully adjusted models the associations between the BBS 3 score and ADL as well as IADL disability were attenuated. This cross-sectional study investigated the associations between balance confidence, 8 functional balance, and physical disability in community-dwelling older men and women 9 who had sustained a previous hip fracture. We found an independent association between 10 decreased balance confidence and ADL/IADL disability as well as between impaired 11 functional balance and ADL/IADL disability. To our knowledge, this is the first study 12 which has simultaneously examined the associations of balance confidence and functional 13 balance with disability in this group of older people with high risk of losing independence. 14 15 Balance confidence and functional balance both have an essential role in coping with daily 16 activities. Moreover, among older hip fracture patients, reduced balance confidence and 17 impaired functional balance may complicate and delay the rehabilitation process. In the 18 present study examining older people with a history of traumatic fall accident, we showed 19 that decreased balance confidence was significantly associated with ADL disability. The 20 association between decreased balance confidence and IADL disability was similar. Our 21 results are in line with previous studies pointing out that fear of falling is associated with 22 increased physical disability in community-dwelling older people [8,11,13]. The majority 23 of hip fractures are a consequence of falls [5,6] and falls with traumatic consequences 24 often generate long-lasting fear of falling [8]. On the other hand, fear of falling leads to activity restriction and exercise avoidance, which in turn results in more severe functional The strengths of the present study include firstly that we recruited a unique clinical group 1 of community-dwelling older people who had sustained a hip fracture. The patient records 2 of the Central Finland central hospital were used for this purpose. Secondly, we included a 3 comprehensive battery of laboratory based physical and functional assessments as well as 4 medical review for health and fracture status. By these assessments we were able to design 5 a valid statistical model, with relevant and necessary confounders, to estimate the 6 association between physical disability, balance, and balance confidence. 7 8 Some study limitations should be noted. The results of this study can't be generalized to all 9 hip fracture patients because the participants in the present study were all relatively healthy. 10 They were all community-dwelling, and those who were unable to walk independently 11 outdoors or had severe diseases or cognitive problems were excluded. Thus, our results 12 probably underestimate the association between functional balance, balance confidence, 13 and physical disability among older hip fracture patients. If we had been able to study also 14 more frail patients, the association would have been even stronger. Additionally, because 15 of the cross-sectional study design, the causality between balance confidence, functional 16 balance, and physical disability remains unclear. Therefore, it is possible that disability 17 caused a sedentary lifestyle and was followed by low balance confidence and impaired 18 functional balance. Either way, there is need for effective physical rehabilitation. 19 20 In conclusion, decreased balance confidence and impaired functional balance are important 21 determinants of physical disability in older people who have sustained a hip fracture. 22 Additionally, the ABC scale complements the functional balance assessment and may 23 potentially be used as a screening tool for impaired functional balance in clinical practice 24 when instant assessment of functional balance is not possible. Furthermore, it is essential to examine what kind of interventions are effective in enhancing functional balance and 1 especially balance confidence among older hip fracture patients. It is also important to 2 study what kind of rehabilitation is effective to reduce physical disability and how older 3 people can be encouraged to be physically active even in the presence of fear of falling, 4 balance impairments, and increased physical disability. In the future, long-enough follow-5 up studies with adequate sample size are needed to assess the effects of different kind of 6 rehabilitation programs as well as the determinants of physical disability after hip fracture. 7 8 9

CONFLICT OF INTEREST STATEMENT 10
The authors of this manuscript have no financial or personal relationships with other 11 people or organizations that could inappropriately bias this work. 12 1 (0-6) Median of the IADL sum score (range) 3 (0-8) All means are expressed: ± SD BBS = Berg Balance Scale ABC = Activities-spesific Balance Confidence Scale ADL = activities of daily living IADL = instrumental activities of daily living